The overriding goal of this project is to assess the cost-effectiveness of health care interventions that improve or prevent deterioration in health related quality of life of the elderly. The absence of well-accepted, validated methods to measure the benefits of improved quality of life often means that such benefits are either undervalued or ignored altogether. We are developing methods to greatly improve the measurement of quality of life changes resulting from the prevention or mitigation of functional limitations. This step may help ensure that health policies place proper value on health interventions that improve the independent functioning of elderly Americans. We plan to use the multimedia preference assessment (utility) software (PALS), developed specifically for computer-inexperienced elders in the current project, ROl AG15110, to estimate the utility for health states of functional impairment of individuals, both with and without functional limitation. We will interview approximately 600 older adults from a random sample stratified by age into two groups: 65 to 74 years, and 75 years or older. We will re-interview approximately 300 of the individuals at one year and two years after their initial interview. We will estimate the impact of an individual's own dependency in Activities of Daily Living (ADLs) on his or her utility rating for heath states of functional dependency. In the longitudinal study, we will define the stability of utilities for health states over time, comparing individuals who do with those who do not themselves develop dependencies during that time. We will assess the impact of emotional and general well-being on utility, both between groups and within individuals over time. Using covariates of age, gender, race/ethnicity, and health status, in addition to the individual's ADL status and emotional well-being, we will estimate a prediction model for utility for health states of functional dependency. Finally, by performing cost-effectiveness evaluation of a geriatric evaluation and management (GEM) intervention, we will illustrate the differences in estimates of cost-effectiveness that can be seen using each of several groups as sources of utility data. The preference assessment methods we are developing will make it possible to conduct formal evaluations of treatments whose primary goal is rehabilitation or preservation of function rather than life prolongation. The longitudinal data from individuals who develop ADL dependencies will deepen our understanding of the impact of functional impairments on older adults.